This proposal for a Mentored Research Scientist Award is designed to provide the candidate with the necessary training and enhanced research experience to transition into an independent drug dependence genetic statistician/genetic epidemiologist. The proposed training, in the areas of drug dependence epidemiology, psychiatric genetics, and kernel-based statistical interaction analysis will complement the candidate's previous training and existing skills, and provide a solid foundation for the candidate to become an independent investigator. This training will be advised by a multi-discipline mentoring team with extensive experience and complementary skills. The Primary Mentor, Dr. Jim Anthony, and Co-Mentor, Dr. Naomi Breslau, will advise the candidate's training in drug dependence epidemiology, with guided learning experiences via consultants on pathophysiology, etiology, clinical diagnosis, and therapeutics. Co-Mentor, Dr. Karen Friderici, and Collaborator, Dr. Laura Bierut, will supervise the candidate's training in psychiatric genetics Co-Mentor, Daniel Schaid, will advise the candidate's training in kernel-based statistical interaction analysis. The long term goal is to identify gene-gene/gene-environment (G-G/G-E) interactions that account for tobacco/nicotine dependence and related phenotypes (NDRP), and then to evaluate their role in NDRP prediction and prevention, and in personalized treatment. The research proposed in this application is to improve the detection of NDRP-related G-G/G-E interactions by employing innovative statistical genetic approaches. In the proposed research, non-parametric kernel-based statistical genetic approaches will be developed and applied to NDRP-associated genes (e.g., nAChRs sub-unit genes) and environmental determinants (e.g., childhood abuse) for (a) detecting high-order interactions, (b) discovering interactions among low-marginal-effect genetic and environmental determinants, and (c) identifying common and unique interactions accounting for NDRP and alcohol dependence and related phenotypes. An existing case-control genome-wide association study dataset from the Study of Addiction: Genetics and Environment will be used for the proposed G-G/G-E interaction analysis. The novel G-G/G-E interaction findings will likely provide new insights into the underlying NDRP pathophysiological and etiological processes, as well as have significant implications for better NDRP prediction and prevention/treatment strategies. The training and research experience gained from the proposed study will serve as the groundwork for an independent research program, including a new R01 proposal to be submitted during the second year of the K01 award interval, in order to strengthen and refine current methods for NDRP-related G-G/G-E interaction research.